The goal of this project is to understand how chronic conduct problems develop from birth through adolescence. Through prospective inquiry, a comprehensive though not exhaustive model will be tested in which biological predispositions (e.g., temperament, heart rate reactivity), sociocultural contexts (e.g., poverty, violent neighborhood), and life experiences (e.g., physical abuse, peer social rejection) are posited as risk factors for conduct problem occurrence and growth. It is proposed that these risk factors combine additively to predict a large portion of variance in adolescent outcomes but also interactively through synergistic effects and moderation of one factor by another. It is hypothesized that life experiences mediate the effects of disposition and context (e.g., difficult temperament and poverty predispose a child to experience harsh parenting which then leads to conduct problem outcomes). Furthermore, this ecosymbiotic developmental model posits reciprocal influences among dispositions, contexts, and life experiences. It is hypothesized that risk factors lead to conduct problems through the development of social knowledge structures which guide cognitive-emotional processes which are proximal to antisocial actions. Developmental sensitivity suggests that different aspects of life experiences are crucial at different developmental eras. This project focuses specifically on adolescent processes, including deviant peer pressure, identity development, and romantic partner relationships. The participants are the 585 boys and girls (100 African-Americans) from three geographic sites and two cohorts of the Child Development Project. They have been followed annually since preschool (age 5) with low attrition, will be in grade 10 (or 9) at the beginning of the proposed project period, and will be followed until the 15th project year 18 months after high school graduation. Data collection will include interviews with parents, adolescents, and their romantic partners, parent-adolescent direct observation, teacher reports, psychiatric interviews, and archival records. Regressions, contrasts of structural equation models, and growth curve plotting will test hypotheses. Emphasis will be given to testing gender-specific and ethnic culture- specific models of antisocial development. The major contributions of this project will be the formulation and empirical testing of a comprehensive theory of how chronic conduct problems develop and indirect implications for universal preventions and preventive interventions for children at high risk for chronic adolescent conduct problems.